3-D Positioning System Based QR Code and Monocular Vision

Abstract

Currently, positioning method based on code can only be realized in 2D space, which cannot be used in non-planar 3D environment. In order to achieve navigation and positioning in a universal 3D environment, this paper designs a monocular visual system to determine position and orientation using QR (quick response) code. This system can measure the 3D position and 3 Euler angles of the monocular camera in a physical coordinate system. The system is composed of two modules: image processing module and pose calculation module. The image processing module performs preprocessing on the obtained image, detecting and sorting of location points, correction and decoding of the QR codes. The pose calculation module first performs the calibration of the internal and external parameters of the camera. Then, using the correspondence between the 2D pixel coordinates and the 3D physical coordinates of the QR codes positioning markers, the relative pose information between the QR code and the camera is solved using efficient perspective-n-point camera pose estimation algorithm to determine the 3D position and posture of the robot and the unmanned vehicle.